# Computations
import numpy as np
import pandas as pd

# scipy
from scipy.stats import norm

# preprocessing
from sklearn import preprocessing
import re

# Visualisation libraries

## Text
from colorama import Fore, Back, Style
from IPython.display import Image, display, Markdown, Latex, clear_output

## progressbar
import progressbar

## plotly
from plotly.offline import init_notebook_mode, iplot 
import plotly.graph_objs as go
import plotly.offline as py
from plotly.subplots import make_subplots
import plotly.express as px

## seaborn
import seaborn as sns

## matplotlib
import matplotlib.pyplot as plt
from matplotlib.patches import Ellipse, Polygon
from matplotlib.font_manager import FontProperties
plt.style.use('seaborn-whitegrid')
plt.rcParams['axes.labelsize'] = 14
plt.rcParams['xtick.labelsize'] = 12
plt.rcParams['ytick.labelsize'] = 12
plt.rcParams['text.color'] = 'k'
%matplotlib inline

## WordCloud
from wordcloud import WordCloud


import warnings
warnings.filterwarnings("ignore")

Home / Business Analytics and Modeling

IBM HR Analytics Employee Attrition and Performance Dataset

• Exploratory Data Analysis
• IBM HR Analytics Modeling using ANN
• IBM HR Analytics Modeling using Stacking Classifiers

IBM HR Analytics Employee Attrition and Performance Dataset

In this study, we analyze HR data available from kaggle.com

This data is fictional and it is created by IBM data scientists.

Categorical Parameters:

	1	2	3	4	5
Education	Below College	College	Bachelor	Master	Doctor
Environment Satisfaction	Low	Medium	High	Very High
Job Involvement	Low	Medium	High	Very High
Job Satisfaction	Low	Medium	High	Very High
Performance Rating	Low	Good	Excellent	Outstanding
Relationship Satisfaction	Low	Medium	High	Very High
WorkLife Balance	Bad	Good	Better	Best

This can be encoded as follows,

Categorical_Dict = {'Education': {1:'Below College', 2:'College',3:'Bachelor', 4: 'Master', 5:'Doctor'},
                    'Environment Satisfaction': {1:'Low', 2:'Medium', 3:'High', 4:'Very High'},
                    'Job Involvement': {1:'Low', 2:'Medium', 3:'High', 4:'Very High'},
                    'Job Satisfaction': {1:'Low', 2:'Medium', 3:'High', 4:'Very High'},
                    'Performance Rating': {1:'Low', 2:'Good', 3:'Excellent', 4:'Outstanding'},
                    'Relationship Satisfaction': {1:'Low', 2:'Medium', 3:'High', 4:'Very High'},
                    'Work Life Balance': {1:'Bad', 2:'Good', 3:'Better', 4:'Best'}}

Loading the Dataset

Path = 'Data/WA_Fn-UseC_-HR-Employee-Attrition.xlsx'
Data = pd.read_excel(Path)
Temp = [re.sub(r"(\w)([A-Z])", r"\1 \2", x) for x in Data.columns]
Temp = [x.replace(' Curr ', ' Current ').replace('18',' 18').replace('Num ','Number Of ') for x in Temp]
Data.columns = Temp
del Temp
Data['Business Travel'] = Data['Business Travel'].str.replace('_',' ')
display(Data.head(8).style.hide_index())
Target = 'Attrition'
Featured_Columns = list(set(Data.columns) - {Target, 'Employee Number'})

Age	Attrition	Business Travel	Daily Rate	Department	Distance From Home	Education	Education Field	Employee Count	Employee Number	Environment Satisfaction	Gender	Hourly Rate	Job Involvement	Job Level	Job Role	Job Satisfaction	Marital Status	Monthly Income	Monthly Rate	Number Of Companies Worked	Over 18	Over Time	Percent Salary Hike	Performance Rating	Relationship Satisfaction	Standard Hours	Stock Option Level	Total Working Years	Training Times Last Year	Work Life Balance	Years At Company	Years In Current Role	Years Since Last Promotion	Years With Current Manager
41	Yes	Travel Rarely	1102	Sales	1	2	Life Sciences	1	1	2	Female	94	3	2	Sales Executive	4	Single	5993	19479	8	Y	Yes	11	3	1	80	0	8	0	1	6	4	0	5
49	No	Travel Frequently	279	Research & Development	8	1	Life Sciences	1	2	3	Male	61	2	2	Research Scientist	2	Married	5130	24907	1	Y	No	23	4	4	80	1	10	3	3	10	7	1	7
37	Yes	Travel Rarely	1373	Research & Development	2	2	Other	1	4	4	Male	92	2	1	Laboratory Technician	3	Single	2090	2396	6	Y	Yes	15	3	2	80	0	7	3	3	0	0	0	0
33	No	Travel Frequently	1392	Research & Development	3	4	Life Sciences	1	5	4	Female	56	3	1	Research Scientist	3	Married	2909	23159	1	Y	Yes	11	3	3	80	0	8	3	3	8	7	3	0
27	No	Travel Rarely	591	Research & Development	2	1	Medical	1	7	1	Male	40	3	1	Laboratory Technician	2	Married	3468	16632	9	Y	No	12	3	4	80	1	6	3	3	2	2	2	2
32	No	Travel Frequently	1005	Research & Development	2	2	Life Sciences	1	8	4	Male	79	3	1	Laboratory Technician	4	Single	3068	11864	0	Y	No	13	3	3	80	0	8	2	2	7	7	3	6
59	No	Travel Rarely	1324	Research & Development	3	3	Medical	1	10	3	Female	81	4	1	Laboratory Technician	1	Married	2670	9964	4	Y	Yes	20	4	1	80	3	12	3	2	1	0	0	0
30	No	Travel Rarely	1358	Research & Development	24	1	Life Sciences	1	11	4	Male	67	3	1	Laboratory Technician	3	Divorced	2693	13335	1	Y	No	22	4	2	80	1	1	2	3	1	0	0	0

First off, let's take a look at the dataset

def Data_Plot(Inp, W = False):
    data_info = Inp.copy()
    data_info = Inp.dtypes.astype(str).to_frame(name='Data Type')
    Temp = Inp.isnull().sum().to_frame(name = 'Number of NaN Values')
    data_info = data_info.join(Temp, how='outer')
    data_info ['Size'] = Inp.shape[0]
    data_info['Percentage'] = 100 - np.round(100*(data_info['Number of NaN Values']/Inp.shape[0]),2)
    data_info = data_info.reset_index(drop = False).rename(columns = {'index':'Features'})
    #
    fig = px.bar(data_info, x= 'Features', y= 'Percentage', color = 'Data Type', text = 'Data Type',
                color_discrete_sequence = ['PaleGreen', 'LightCyan', 'PeachPuff', 'Pink', 'Plum'],
                 hover_data = data_info.columns)
    fig.update_layout(plot_bgcolor= 'white', legend=dict(x=1.01, y=.5, traceorder="normal",
                                                         bordercolor="DarkGray", borderwidth=1))
    fig.update_traces(texttemplate= 6*' ' + '%{label}', textposition='inside')
    fig.update_traces(marker_line_color= 'Black', marker_line_width=1., opacity=1)
    if W:
        fig.update_layout(width = W)
    fig.show()
    return data_info
    
    
_ = Data_Plot(Data)

Moreover,

def Distinct_Observations(Inp, Target = Target, Featured_Columns = None, YL = None):
    if Featured_Columns == None:
        Featured_Columns = list(set(Data.columns) - {Target})
    Temp = Inp[Featured_Columns].nunique()
    fig = go.Figure([go.Bar(x=Temp.index, y=Temp.values)])
    fig.update_traces(marker_line_color= 'Navy', marker_line_width=1, opacity=1, showlegend = False)
    fig.update_layout(legend_orientation='v', plot_bgcolor= 'white', height= 450, width= 980,
                     title={'text': '<b>' + 'Distinct Observations in Each Column' + '<b>', 'x':0.5,
                            'y': 0.92, 'xanchor': 'center', 'yanchor': 'top'},
                     yaxis_title='Frequency')
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                         zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                         showgrid=False, gridwidth=1, gridcolor='Lightgray')
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=True, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray')
    if not YL == None:
        fig.update_yaxes(range =[0, YL])
    fig.show()
    return Temp
Temp = Distinct_Observations(Inp = Data, Featured_Columns = Featured_Columns, YL = 1500)
Temp.loc[Temp>5].sort_index().to_frame('Distinct Observations')

	Distinct Observations
Age	43
Daily Rate	886
Distance From Home	29
Education Field	6
Hourly Rate	71
Job Role	9
Monthly Income	1349
Monthly Rate	1427
Number Of Companies Worked	10
Percent Salary Hike	15
Total Working Years	40
Training Times Last Year	7
Years At Company	37
Years In Current Role	19
Years Since Last Promotion	16
Years With Current Manager	18

Exploratory Data Analysis

## Attrition Colormap
Att_Colors = ['LightSalmon', 'LightBlue']
Att_LC = 'Black'
# Gender Colormap
MF_Colors = ['HotPink', 'RoyalBlue']
MF_LC = 'Navy'
# Education
Ed_Colors = ['LightCoral','Khaki','GreenYellow','LimeGreen','ForestGreen']
Ed_LC = 'Black'

def DistPlot(Feat, yLim = [0, 80], H = 450, titleY = 0.92):
    fig = px.histogram(Data, x = Feat, color='Attrition', marginal= 'box',
                       color_discrete_sequence= Att_Colors, hover_data=Data.columns)
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=False, gridwidth=1, gridcolor='Lightgray')
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=True, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray')
    Name = '%s Distribution by Attrition' % Feat
    fig.update_layout(legend_orientation='v', plot_bgcolor= 'white', height= H, width= 980,
                     title={'text': '<b>' + Name + '<b>', 'x':0.5, 'y': titleY, 'xanchor': 'center', 'yanchor': 'top'},
                     yaxis_title='Frequency')
    fig.update_traces(marker_line_color= Att_LC, marker_line_width=0.5, opacity=1)
    fig['layout']['yaxis'].update(range=yLim)
    fig.show()
    
def PlotX1(df, Feat, ColorFeat = 'Gender', yLim = [0, 35], H = 500, titleY = 0.90):
    fig = make_subplots(rows=1, cols=2, horizontal_spacing = 0.02, shared_yaxes=True,
#                         x_title = Feat,
                        y_title = 'Percent',
                        subplot_titles=('Attrition: No', 'Attrition: Yes'))
    # Left
    fig1 = px.bar(df.loc[df.Attrition == 'No'], x=  Feat, y= 'Percentage', orientation='v',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns,
                  color_discrete_sequence = MF_Colors)

    for i in range(len(fig1['data'])):
        fig.add_trace(fig1['data'][i], row=1, col=1)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, row=1, col=1)

    # Right
    fig2 = px.bar(df.loc[df.Attrition == 'Yes'], x= Feat, y= 'Percentage', orientation='v',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns,
                  color_discrete_sequence = MF_Colors)

    for i in range(len(fig2['data'])):
        fig.add_trace(fig2['data'][i], row=1, col=2)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, showlegend = False, row=1, col=2)

    # Update
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=False, gridwidth=1, gridcolor='Lightgray')
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray', range= yLim)
    fig.update_layout(legend_orientation='v', legend_title_text=ColorFeat, plot_bgcolor= 'white', height= H, width= 980)
    fig.update_layout(legend=dict(font=dict(color="Black"), bordercolor="Lightgray", borderwidth=1))
    Name = '%s Distribution by %s and Attrition' % (Feat, ColorFeat)
    fig.update_layout(title={'text': '<b>' + Name + '<b>', 'x':0.5, 'y': titleY, 'xanchor': 'center', 'yanchor': 'top'})
    fig.show() 
    
def PlotX2(df, Feat, ColorFeat = 'Education', yLim = [0, 35], H = 500, titleY = 0.90):
    fig = make_subplots(rows=1, cols=2, horizontal_spacing = 0.02, shared_yaxes=True,
#                         x_title = Feat,
                        y_title = 'Percent',
                        subplot_titles=('Attrition: No', 'Attrition: Yes'))
    # Left
    fig1 = px.bar(df.loc[df.Attrition == 'No'], x= Feat, y= 'Percentage', orientation='v', barmode='group',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns, 
                  color_discrete_sequence = Ed_Colors)
    for i in range(len(fig1['data'])):
        fig.add_trace(fig1['data'][i], row=1, col=1)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, row=1, col=1)

    # Right
    fig2 = px.bar(df.loc[df.Attrition == 'Yes'], x= Feat, y= 'Percentage', orientation='v',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns,
                  color_discrete_sequence = Ed_Colors)

    for i in range(len(fig2['data'])):
        fig.add_trace(fig2['data'][i], row=1, col=2)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, showlegend = False, row=1, col=2)

    # Update
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=False, gridwidth=1, gridcolor='Lightgray')
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray', range= yLim)
    fig.update_layout(legend_orientation='v', legend_title_text=ColorFeat, plot_bgcolor= 'white', height= H, width= 980)
    fig.update_layout(legend=dict(font=dict(color="Black"), bordercolor="Lightgray", borderwidth=1))
    Name = '%s Distribution by %s and Attrition' % (Feat, ColorFeat)
    fig.update_layout(title={'text': '<b>' + Name + '<b>', 'x':0.5, 'y': titleY, 'xanchor': 'center', 'yanchor': 'top'})
    fig.show()   

def PlotY1(df, Feat, ColorFeat = 'Gender', xLim = [0, 35], H = 500, titleY = 0.90):
    fig = make_subplots(rows=2, cols=1, vertical_spacing = 0.08, shared_yaxes=True,
#                         y_title = Feat,
                        x_title = 'Percent',
                        subplot_titles=('Attrition: No', 'Attrition: Yes'))
    # Top
    fig1 = px.bar(df.loc[df.Attrition == 'No'], y=  Feat, x= 'Percentage', orientation='h',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns,
                  color_discrete_sequence = MF_Colors)

    for i in range(len(fig1['data'])):
        fig.add_trace(fig1['data'][i], row=1, col=1)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, row=1, col=1)

    # Bottom
    fig2 = px.bar(df.loc[df.Attrition == 'Yes'], y= Feat, x= 'Percentage', orientation='h',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns,
                  color_discrete_sequence = MF_Colors)

    for i in range(len(fig2['data'])):
        fig.add_trace(fig2['data'][i], row=2, col=1)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, showlegend = False, row=2, col=1)

    # Update
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray', range= xLim)
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=False, gridwidth=1, gridcolor='Lightgray')
    fig.update_layout(legend_orientation='v', legend_title_text=ColorFeat, plot_bgcolor= 'white', height= H, width= 980)
    fig.update_layout(legend=dict(font=dict(color="Black"), bordercolor="Lightgray", borderwidth=1))
    Name = '%s Distribution by %s and Attrition' % (Feat, ColorFeat)
    fig.update_layout(title={'text': '<b>' + Name + '<b>', 'x':0.5, 'y': titleY, 'xanchor': 'center', 'yanchor': 'top'})
    fig.show() 
    
def PlotY2(df, Feat, ColorFeat = 'Education', xLim = [0, 35], H = 500, titleY = 0.90):
    fig = make_subplots(rows=2, cols=1, vertical_spacing = 0.08, shared_yaxes=True,
#                         y_title = Feat,
                        x_title = 'Percent',
                        subplot_titles=('Attrition: No', 'Attrition: Yes'))
    # Top
    fig1 = px.bar(df.loc[df.Attrition == 'No'], y= Feat, x= 'Percentage', orientation='h', barmode='group',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns, 
                  color_discrete_sequence = Ed_Colors)
    for i in range(len(fig1['data'])):
        fig.add_trace(fig1['data'][i], row=1, col=1)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, row=1, col=1)

    # Bottom
    fig2 = px.bar(df.loc[df.Attrition == 'Yes'], y= Feat, x= 'Percentage', orientation='h',
                  color = ColorFeat, text = 'Percentage', hover_data= df.columns,
                  color_discrete_sequence = Ed_Colors)

    for i in range(len(fig2['data'])):
        fig.add_trace(fig2['data'][i], row=2, col=1)
    fig.update_traces(marker_line_color= MF_LC, marker_line_width=1, opacity=1, showlegend = False, row=2, col=1)

    # Update
    fig.update_xaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=True, gridwidth=1, gridcolor='Lightgray', range= xLim)
    fig.update_yaxes(showline=True, linewidth=1, linecolor='Lightgray', mirror=True,
                     zeroline=False, zerolinewidth=1, zerolinecolor='Black',
                     showgrid=False, gridwidth=1, gridcolor='Lightgray')
    fig.update_layout(legend_orientation='v', legend_title_text=ColorFeat, plot_bgcolor= 'white', height= H, width= 980)
    fig.update_layout(legend=dict(font=dict(color="Black"), bordercolor="Lightgray", borderwidth=1))
    Name = '%s Distribution by %s and Attrition' % (Feat, ColorFeat)
    fig.update_layout(title={'text': '<b>' + Name + '<b>', 'x':0.5, 'y': titleY, 'xanchor': 'center', 'yanchor': 'top'})
    fig.show()

# For plotting
def FeatCut(Feat, Bins, ColorFeat = 'Gender', Inp = Data):
    Out = Inp[[Feat, ColorFeat,'Attrition']]
    Out[Feat] = pd.cut(Out[Feat], bins = pd.IntervalIndex.from_tuples([(x, y) for x, y in zip(Bins[:-1],Bins[1:])]))

    Out = Out.groupby([Feat, ColorFeat,'Attrition'])['Attrition'].agg({'count'}).rename(columns = {'count':'Count'})
    Out['Percentage'] = np.round(100* Out.values /Out.sum().values, 2)
    Out.reset_index(drop = False, inplace = True)
    Out = Out.sort_values(by=[Feat])
    Out[Feat] = Out[Feat].astype(str)
    return Out

def Table(Feat, ColorFeat = 'Gender', Inp = Data):
    Out = Inp[[Feat, ColorFeat,'Attrition']]
    Out = Out.groupby([Feat, ColorFeat,'Attrition'])['Attrition'].agg({'count'}).rename(columns = {'count':'Count'})
    Out['Percentage'] = np.round(100* Out.values /Out.sum().values, 2)
    Out.reset_index(drop = False, inplace = True)
    Out = Out.sort_values(by=[Feat])
    Out[Feat] = Out[Feat].astype(str)
    return Out

# For preprocessing
def ColBins(Inp, Bins):
    Out = pd.cut(Inp, bins = pd.IntervalIndex.from_tuples([(x, y) for x, y in zip(Bins[:-1],Bins[1:])]))
    Temp = np.sort(Out.astype('str').unique())
    Dict = dict(zip(Temp, np.arange(len(Temp))))
    Out = Out.astype('str').replace(Dict)
    return Out

# A copy of the dataset
df = Data.copy()
# Modifying the dataset for plotting only
for Feat in Categorical_Dict.keys():
    Data [Feat] = Data[Feat].replace(Categorical_Dict[Feat])
del Feat

Age

Feat = 'Age'
Bins = [15, 24, 40, 59, 80]
DistPlot(Feat)
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat)
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 20])
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Feat, Bins

Business Travel

Feat = 'Business Travel'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 40], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 25], H = 500, titleY = 0.90)
del Feat

Daily Rate

Feat = 'Daily Rate'
Bins = [100, 300, 600, 1000, 1500]
DistPlot(Feat, yLim = [0, 250], H = 450, titleY = 0.92)
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat, yLim = [0, 20])
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 12])
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Feat

Department

Feat = 'Department'
PlotY1(df = Table(Feat), Feat = Feat, xLim = [0, 40], H = 550, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 25])
del Feat

Distance from Home

Feat = 'Distance From Home'
Bins = [0, 5, 10, 20, 30]
DistPlot(Feat, yLim = [0, 250], H = 450, titleY = 0.92)
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat, yLim = [0, 30])
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 20])
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Feat

Education

Feat = 'Education'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 25])
del Feat

Education Field

Feat = 'Education Field'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 25])
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 14])
del Feat

Environment Satisfaction

Feat = 'Environment Satisfaction'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 20])
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 12])
del Feat

Hourly Rate

Feat = 'Hourly Rate'
Bins = [25, 50, 75, 101]
DistPlot(Feat, yLim = [0, 140], H = 500, titleY = 0.92)
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 14], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Bins, Feat

Job Involvement

Feat = 'Job Involvement'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 35], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
del Feat

Job Level

Feat = 'Job Level'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 12], H = 500, titleY = 0.90)
del Feat

Job Roles

Feat = 'Job Role'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 12], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 8], H = 500, titleY = 0.90)
del Feat

Job Satisfaction

Feat = 'Job Satisfaction'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 10], H = 500, titleY = 0.90)
del Feat

Marital Status

Feat = 'Marital Status'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 25], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 16], H = 500, titleY = 0.90)
del Feat

Monthly Income

Feat = 'Monthly Income'
Bins = [1e3, 3e3, 7e3, 1e4, 2e4]
Bins = [int(x) for x in Bins]
DistPlot(Feat, yLim = [0, 400], H = 500, titleY = 0.92)
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat, yLim = [0, 25], H = 500, titleY = 0.90)
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 14], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Bins, Feat

Monthly Rate

Feat = 'Monthly Rate'
Bins = [2e3, 1e4, 2e4, 3e4]
Bins = [int(x) for x in Bins]
DistPlot(Feat, yLim = [0, 80], H = 500, titleY = 0.92)
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 14], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Bins, Feat

Number of Companies Worked

Feat = 'Number Of Companies Worked'
Bins = [-1, 2, 5, 10]
DistPlot(Feat = Feat, yLim = [0, 600], H = 450, titleY = 0.92)
Temp = FeatCut(Feat, Bins = Bins).replace({'(-1, 2]': '[0, 2]'})
PlotX1(df = Temp, Feat = Feat, yLim = [0, 35], H = 500, titleY = 0.90)
Temp = FeatCut(Feat, Bins, ColorFeat = 'Education').replace({'(-1, 2]': '[0, 2]'})
PlotX2(df = Temp, Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Temp, Feat

Over Time

Feat = 'Over Time'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 40], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 30], H = 500, titleY = 0.90)
del Feat

Percent Salary Hike

Bins = list(np.arange(10, 27, 4))
Feat = 'Percent Salary Hike'
PlotX1(df = FeatCut(Feat, Bins = Bins), Feat = Feat, yLim = [0, 30], H = 500, titleY = 0.90)
PlotX2(df = FeatCut(Feat, Bins, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Feat

Performance Rating

Feat = 'Performance Rating'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 50], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 30], H = 500, titleY = 0.90)
del Feat

Relationship Satisfaction

Feat = 'Relationship Satisfaction'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 12], H = 500, titleY = 0.90)
del Feat

Stock Option Level

Feat = 'Stock Option Level'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 25], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 16], H = 500, titleY = 0.90)
del Feat

Total Working Years

Feat = 'Total Working Years'
Bins = [-1, 10, 20, 30, 41]
DistPlot(Feat = Feat, yLim = [0, 250], H = 450, titleY = 0.92)
Temp = FeatCut(Feat, Bins = Bins).replace({'(-1, 10]': '[0, 10]'})
PlotX1(df = Temp, Feat = Feat, yLim = [0, 35], H = 500, titleY = 0.90)
Temp = FeatCut(Feat, Bins, ColorFeat = 'Education').replace({'(-1, 10]': '[0, 10]'})
PlotX2(df = Temp, Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Temp, Feat

Training Times Last Year

Feat = 'Training Times Last Year'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 12], H = 500, titleY = 0.90)
del Feat

Work-Life Balance Score

Feat = 'Work Life Balance'
PlotX1(df = Table(Feat), Feat = Feat, yLim = [0, 35], H = 500, titleY = 0.90)
PlotX2(df = Table(Feat, ColorFeat = 'Education'), Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
del Feat

Years at the Company

Feat = 'Years At Company'
Bins = [-1, 10, 20, 30, 41]
DistPlot(Feat, yLim = [0, 200], H = 450, titleY = 0.92)
Temp = FeatCut(Feat, Bins = Bins).replace({'(-1, 10]': '[0, 10]'})
PlotX1(df = Temp, Feat = Feat, yLim = [0, 50], H = 500, titleY = 0.90)
Temp = FeatCut(Feat, Bins, ColorFeat = 'Education').replace({'(-1, 10]': '[0, 10]'})
PlotX2(df = Temp, Feat = Feat, yLim = [0, 30], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Temp, Feat

Years In Current Role

Feat = 'Years In Current Role'
Bins = [-1, 4, 8, 12, 19]
DistPlot(Feat = Feat, yLim = [0, 400], H = 450, titleY = 0.92)
Temp = FeatCut(Feat, Bins = Bins).replace({'(-1, 4]': '[0, 4]'})
PlotX1(df = Temp, Feat = Feat, yLim = [0, 30], H = 500, titleY = 0.90)
Temp = FeatCut(Feat, Bins, ColorFeat = 'Education').replace({'(-1, 4]': '[0, 4]'})
PlotX2(df = Temp, Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Temp, Feat

Years Since Last Promotion

Feat = 'Years Since Last Promotion'
Bins = [-1, 5, 10, 16]
DistPlot(Feat = Feat, yLim = [0, 600], H = 450, titleY = 0.92)
Temp = FeatCut(Feat, Bins = Bins).replace({'(-1, 5]': '[0, 5]'})
PlotX1(df = Temp, Feat = Feat, yLim = [0, 50], H = 500, titleY = 0.90)
Temp = FeatCut(Feat, Bins, ColorFeat = 'Education').replace({'(-1, 5]': '[0, 5]'})
PlotX2(df = Temp, Feat = Feat, yLim = [0, 30], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Temp, Feat

Years With Current Manager

Feat = 'Years With Current Manager'
Bins = [-1, 4, 8, 12, 18]
DistPlot(Feat = Feat, yLim = [0, 400], H = 450, titleY = 0.92)
Temp = FeatCut(Feat, Bins = Bins).replace({'(-1, 4]': '[0, 4]'})
PlotX1(df = Temp, Feat = Feat, yLim = [0, 35], H = 500, titleY = 0.90)
Temp = FeatCut(Feat, Bins, ColorFeat = 'Education').replace({'(-1, 4]': '[0, 4]'})
PlotX2(df = Temp, Feat = Feat, yLim = [0, 20], H = 500, titleY = 0.90)
# For Preprocessing
df[Feat] = ColBins(Inp = df[Feat], Bins = Bins)
del Temp, Feat

Now

_ = Data_Plot(Data)
_ = Distinct_Observations(Inp = df, Featured_Columns = Featured_Columns, YL = 10)

Problem Description

In the dataset, Attrition represents whether an employee is churned or not. We would like to create a predictive model that predicts this feature.

We need to convert categorical data to numeric data.

def dtypes_group(Inp):
    Temp = Inp.dtypes.to_frame(name='Data Type').sort_values(by=['Data Type'])
    Out = pd.DataFrame(index =Temp['Data Type'].unique(), columns = ['Features','Count'])
    for c in Temp['Data Type'].unique():
        Out.loc[Out.index == c, 'Features'] = [Temp.loc[Temp['Data Type'] == c].index.tolist()]
        Out.loc[Out.index == c, 'Count'] = len(Temp.loc[Temp['Data Type'] == c].index.tolist())
    Out = Out.reset_index(drop = False).rename(columns = {'index':'Data Type'})
    Out['Data Type'] = Out['Data Type'].astype(str)
    return Out

def dtype_sep(Inp):
    Temp = Inp.dtypes.reset_index(drop = False)
    Temp.columns = ['Features', 'Data Type']
    Temp['Data Type'] = Temp['Data Type'].astype(str)
    # Numeric_Columns
    Numeric_Columns = Temp.loc[Temp['Data Type'].isin(['int64', 'int32', 'float64', 'float32']),'Features'].tolist()
    # Categorical_Columns
    Categorical_Columns = Temp.loc[Temp['Data Type'] == 'object','Features'].tolist()
    return Numeric_Columns, Categorical_Columns

Numeric_Columns, Categorical_Columns = dtype_sep(Data)
display(dtypes_group(Data).style.hide_index())

Data Type	Features	Count
int64	['Age', 'Years In Current Role', 'Years At Company', 'Training Times Last Year', 'Total Working Years', 'Stock Option Level', 'Standard Hours', 'Percent Salary Hike', 'Number Of Companies Worked', 'Monthly Rate', 'Monthly Income', 'Years Since Last Promotion', 'Job Level', 'Years With Current Manager', 'Hourly Rate', 'Daily Rate', 'Employee Number', 'Employee Count', 'Distance From Home']	19
object	['Attrition', 'Business Travel', 'Work Life Balance', 'Department', 'Education', 'Relationship Satisfaction', 'Performance Rating', 'Education Field', 'Over Time', 'Over 18', 'Environment Satisfaction', 'Gender', 'Job Satisfaction', 'Job Role', 'Job Involvement', 'Marital Status']	16

We can use LabelEncoder for converting categorical to numeric using. Therefore,

N = len(Categorical_Columns)
# Progressbar
Counter = 0
Progress_Bar = progressbar.ProgressBar(maxval= N, widgets=[progressbar.Bar('=', '|', '|'), progressbar.Percentage()])

#--------------- the loop ----------------------
Progress_Bar.start()
for i in range(N):
    le = preprocessing.LabelEncoder()
    le.fit(list(df[Categorical_Columns[i]]))
    df[Categorical_Columns[i]] = le.transform(df[Categorical_Columns[i]])
    del le
    Progress_Bar.update(Counter)
    Counter+=1
Progress_Bar.finish()
#--------------- End of the loop ---------------

# Finally, converting values of df back to integers.
df = df.astype(int)

display(dtypes_group(df).style.hide_index())

|=========================================================================|100%

Data Type	Features	Count
int32	['Age', 'Monthly Rate', 'Number Of Companies Worked', 'Over 18', 'Over Time', 'Percent Salary Hike', 'Performance Rating', 'Monthly Income', 'Relationship Satisfaction', 'Stock Option Level', 'Total Working Years', 'Training Times Last Year', 'Work Life Balance', 'Years At Company', 'Years In Current Role', 'Standard Hours', 'Years Since Last Promotion', 'Marital Status', 'Job Role', 'Attrition', 'Business Travel', 'Daily Rate', 'Department', 'Distance From Home', 'Education', 'Job Satisfaction', 'Education Field', 'Employee Number', 'Environment Satisfaction', 'Gender', 'Hourly Rate', 'Job Involvement', 'Job Level', 'Employee Count', 'Years With Current Manager']	35

Variance of the Features

Features with variance zero

First, we remove features that have zero variance as these features don't add anything to our modeling.

# var
Temp = df[Featured_Columns].var().sort_values(ascending = False)
Temp = Temp.loc[Temp.round(16) ==0].index.tolist()
print(Back.BLACK + Fore.CYAN + Style.NORMAL + 'Features with variance zero' + Style.RESET_ALL + ':' + '%s' % ', '.join(Temp))
df = df.drop(columns = Temp)
del Temp

Features with variance zero:Employee Count, Over 18, Standard Hours

X and y sets

Aditional_Columns = [Target, 'Employee Number']
X = df.drop(columns = Aditional_Columns)
y = df[Target]

Features with high variance

Moreover, high variance for some features can hurt our modeling process. For this reason, we would like to standardize features by removing the mean and scaling to unit variance. In this article, we demonstrated the benefits of scaling data using StandardScaler().

# scaling data
scaler = preprocessing.StandardScaler()
X_std = scaler.fit_transform(X)
X_std = pd.DataFrame(data = X_std, columns =X.columns)
del scaler

fig, ax = plt.subplots(2, 1, figsize=(18, 8))
ax = ax.ravel()
font = FontProperties()
font.set_weight('bold')
CP = [sns.color_palette("OrRd", 20), sns.color_palette("Greens", 20)]
Names = ['Variance of the Features', 'Variance of the Features (Standardized)']
Sets = [X, X_std]
kws = dict(label='Feature\nVariance', aspect=20, shrink= .3)
for i in range(len(ax)):
    Temp = Sets[i].var().sort_values(ascending = False).to_frame(name= 'Variance').round(2).T
    _ = sns.heatmap(Temp, ax=ax[i], annot=True, square=True,  cmap = CP[i],
                    linewidths = 0.8, vmin=0, vmax=Temp.max(axis =1)[0],  annot_kws={"size": 6},
                    cbar_kws=kws)
    _ = ax[i].set_yticklabels('')
    _ = ax[i].set_title(Names[i], fontproperties=font, fontsize = 16)
    del Temp
del CP, Names, ax, fig, font, Sets, kws

Modifying dataset.

df[X.columns.tolist()] = X_std[X.columns.tolist()]

Saving to a CSV

df.to_csv (Path.split(".")[0]+'_STD.csv', index = None, header=True)

---

References

1. Kaggle Dataset: IBM HR Analytics Employee Attrition & Performance
2. Getting Started with Plotly in Python

---